Max Geissbuhler, Slingshot Aerospace; Joshua Horwood, Slingshot Aerospace; Todd Brost, Slingshot Aerospace; Navraj Singh, Slingshot Aerospace; Jeff Aristoff, Slingshot Aerospace
Keywords: Conjunction Assessment, Opportunity Screening, Orbit Propagation, Orbit Determination, Maneuver Estimation, Commercial Data, Space Situational/Domain Awareness, Rendezvous Proximity Operation
Abstract:
At AMOS 2016, Numerica (recently acquired by Slingshot Aerospace) presented a novel approach to conjunction assessment (CA) screening and probability of collision calculation [1]. Like other conjunction analysis tools in existence today, the software implementation of this approach, called KRATOS (Kollision Risk Assessment Tool in Orbital element Spaces), originally only addressed on-orbit hazards (i.e., collisions with benign payloads or space debris). However, in the high-stakes space domain, additional considerations are needed to address on-orbit threats. First, one must consider the maneuvering capabilities of the conjuncting spacecraft. Second, one must consider other factors such as lighting conditions and relative velocities. It is thus necessary to screen for opportunistic conjunctions which we call “opportunistic conjunction screening,” or more succinctly, “opportunity screening.”
To address this need, Slingshot has developed an innovative opportunity screening tool in the form of a web application that is used to provide key insights to its space operations customers. Specifically, the opportunity screening tool performs an “all-vs-all” CA screening using KRATOS where the “all” comprises all active payloads. A higher-than-normal miss distance threshold than what would normally be used in a traditional CA run is used to find all possible screening opportunities (e.g., a primary that may attempt to image a secondary). Conjunctions are further filtered to include only those that favor screening opportunities, e.g., a low relative velocity at the time-of-closest approach and favorable lighting conditions inferred from solar and lunar phase angles. The web app presents those screening opportunities to the analyst in the form of a sortable and filterable table. The user may select a particular conjunction and choose to examine additional metrics and graphics on the details of the screening opportunity.
Importantly, the opportunity screening tool also accounts for possible maneuver plans and thrusting capabilities of any of the active payloads used in the CA. In other words, the prediction of orbital states used in the CA assumes not only nominal motion (gravity, third-body perturbations, atmospheric drag, solar radiation pressure, etc.), but also permits maneuver plans and estimated or prescribed thrusting capabilities in the orbital predictions. A use case is when an adversarial spacecraft is actively thrusting, possibly to approach another spacecraft to perform imaging or some other nefarious action, and one would like to know which spacecrafts could be targets over a particular screening window.
In this paper, we present two main algorithmic components that are used in the latest version of the opportunity screening web app that account for maneuvering spacecraft and support the aforementioned use case. The first component is an enhancement to Slingshot’s Implicit Runge-Kutta orbital propagator (IRKProp) [2,3] that permits the modeling of a continuous thrust acceleration (parameterized in terms of the thrusting start and end times, total velocity increment (Delta-v) magnitude, propellant exhaust velocity magnitude, and thrusting direction vector) that is sufficiently general to accommodate different types of thrusters such as chemical, ion, and electric. The second component is a new batch least squares estimator that jointly estimates an object’s position-velocity state, drag and solar radiation pressure ballistic coefficients, and thrusting parameters given a sequence of angle-only observations from one or more optical sensors.
Besides presenting the underlying optimization problem for joint orbital state and maneuver estimation, we demonstrate that its solution is feasible and accurate for several reasons. First, we can “hot start” the iterative non-linear least squares procedure using an approximate orbital state and maneuver alert generated from Slingshot’s Multiple Frame Assignment Space Tracker (MFAST) software [4]. Second, we can obtain sufficient data needed to estimate all orbital state components and thrusting parameters from sensors in the Slingshot Global Sensor Network (GSN), even when a spacecraft is actively thrusting. Third, the accuracy of the orbital states generated from Slingshot’s IRKProp is comparable to other Special Perturbations-based orbital propagators. The results from a study that compare the accuracy of Slingshot’s IRKProp against publicly available high accuracy ephemerides are presented in this paper. We also show results from applying the new orbital state-thrusting parameter batch estimator using real GSN data on both active LEO and GEO objects. Ultimately, we demonstrate that once an accurate orbital state is estimated with thrusting parameters, one can accurately propagate the orbit and assess screening opportunities against other objects via KRATOS, which has been upgraded to ingest maneuver plans.
References
[1] J. T. Horwood, N. Singh, J. M. Aristoff, and A. Bhopale, “KRATOS: Kollision Risk Assessment Tool in Orbital element Spaces,” in Proceedings of the 2016 Advanced Maui Optical and Space Surveillance Technologies Conference, Wailea, HI, September 2016.
[2] J. M. Aristoff, J. T. Horwood, and A. B. Poore, “Orbit and uncertainty propagation: a comparison of Gauss–Legendre-, Dormand–Prince-, and Chebyshev–Picard-based approaches,” Celestial Mechanics and Dynamical Astronomy, vol. 117, pp. 13-28, 2013.
[3] J. M. Aristoff, J. T. Horwood, and A. B. Poore, “Implicit Runge-Kutta-based methods for fast, precise, and scalable uncertainty propagation,” Celestial Mechanics and Dynamical Astronomy, vol. 122, no. 2, pp. 169-182, 2015.
[4] J. M. Aristoff, D. J. C. Beach, P. A. Ferris, J. T. Horwood, A. D. Mont, N. Singh, and A. B. Poore, “Multiple Frame Assignment Space Tracker (MFAST): results on UCT processing,” in Proceedings of the 2015 AIAA/AAS Astrodynamics Specialist Conference, Vail, CO, August 2015 (Paper AAS-15-675).
Date of Conference: September 27-20, 2022
Track: Conjunction/RPO